ktnyt/cclsp

• highreadme#1

  Reposition README H1 and opening paragraph to emphasize LLM-LSP integration

  原因：

  当前

  # cclsp - not your average LSP adapter
  
  **cclsp** is a Model Context Protocol (MCP) server that seamlessly integrates LLM-based coding agents with Language Server Protocol (LSP) servers. LLM-based coding agents often struggle with providing accurate line/column numbers, which makes naive attempts to integrate with LSP servers fragile and frustrating. cclsp solves this by intelligently trying multiple position combinations and providing robust symbol resolution that just works, no matter how your AI assistant counts lines.

  复制粘贴的修复

  复制

  # cclsp: The Claude Code LSP Adapter for Robust LLM-LSP Integration
  
  **cclsp** is a Model Context Protocol (MCP) server designed to seamlessly integrate LLM-based coding agents, such as Claude Code, with any Language Server Protocol (LSP) server. It specifically solves the common problem of LLM-based agents struggling with accurate line/column numbers, providing robust symbol resolution that just works, regardless of how your AI assistant counts lines.

  待办完成忽略
• mediumtopics#2

  Add more specific AI/LLM integration topics

  原因：

  当前

  claude, claude-code, lsp, mcp, mcp-server

  复制粘贴的修复

  复制

  claude, claude-code, lsp, mcp, mcp-server, ai-coding-assistant, llm-integration, code-agent-integration

  待办完成忽略
• mediumreadme#3

  Add a 'Comparison' or 'What it's not' section to clarify unique positioning

  原因：

  复制粘贴的修复

  复制

  ## What cclsp is (and isn't)
  cclsp is not a language-specific LSP server (like `rust-analyzer` or `pyright`), nor is it an AI coding agent itself (like `GitHub Copilot` or `Code Llama`). Instead, cclsp acts as a crucial adapter, enabling *any* LLM-based coding agent to reliably interact with *any* existing LSP server, specifically by resolving the common issue of inaccurate line/column number reporting from AI models.

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How to integrate AI coding assistants with existing Language Server Protocol implementations?

  你：未被推荐
  AI 推荐顺序：

  1. Copilot.vim (github/copilot.vim)
  2. Copilot.nvim (github/copilot.nvim)
  3. coc.nvim (neoclide/coc.nvim)
  4. nvim-lspconfig (neovim/nvim-lspconfig)
  5. GitHub Copilot
  6. Codeium
  7. Tabnine (tabnine/Tabnine-VSCode)
  8. IntelliJ IDEA
  9. AI Assistant
  10. PyCharm
  11. rust-analyzer (rust-lang/rust-analyzer)
  12. pyright (microsoft/pyright)
  13. Tabby (TabbyML/tabby)
  14. Code Llama

  AI 推荐了 14 个替代方案，却始终没点名 ktnyt/cclsp。这就是要补上的差距。

  查看 AI 完整回答
• 品类问题

  Seeking a solution for LLM code agents providing inaccurate line numbers to LSP.

  你：未被推荐
  AI 推荐顺序：

  1. Neovim (neovim/neovim)
  2. nvim-treesitter (nvim-treesitter/nvim-treesitter)
  3. VS Code (microsoft/vscode)
  4. Tree-sitter (tree-sitter/tree-sitter)
  5. Pyright (microsoft/pyright)
  6. rust-analyzer (rust-lang/rust-analyzer)
  7. typescript-language-server (typescript-language-server/typescript-language-server)
  8. ast
  9. syn (dtolnay/syn)
  10. Babel (babel/babel)
  11. TypeScript's own compiler API
  12. Sourcegraph (sourcegraph/sourcegraph)
  13. ctags (universal-ctags/ctags)
  14. cscope (cscope-devs/cscope)
  15. Llama 2
  16. Code Llama
  17. GPT-3.5
  18. GPT-4

  AI 推荐了 18 个替代方案，却始终没点名 ktnyt/cclsp。这就是要补上的差距。

  查看 AI 完整回答

• Metadata completeness

  pass

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of ktnyt/cclsp?

  pass
  ✓AI 明确点名了 ktnyt/cclsp

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts ktnyt/cclsp in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 ktnyt/cclsp

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo ktnyt/cclsp solve, and who is the primary audience?

  pass
  ✓AI 明确点名了 ktnyt/cclsp

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？